12:35 UT Good news! The 27-m receiver is back. When installed, it seemed to have a problem still with the Hi H polarization (0 drain current), but this morning the drain current is back... Mark says he will try replacing the Bias board today, but in the meantime, I am taking data on 2136+006, with pcal_hi-all.fsq, with the 200 MHz correlator. I hope to be able to set the delays as soon as the first data are available (the first scan started 12:34 UT). Note that I also updated the FEM and DCM attenuation settings this morning, so everything should be more or less operational.
13:15 UT The delays were very bad! But I was able to find relatively good settings, and have updated the delays and restarted the scan. I restarted it at 13:13 UT, but then aborted, increased the time on source by another hour, and restarted at 13:15 UT. So the actual useful scan will start at 13:15 UT and go until 14:34 UT. I hope there are no flares that we'll miss in that extra hour...there is a little flare going on right now, but too early for us anyway.
10:45 UT We should have two days of reasonable data for calibrating (full-disk, but no flares), on Jun 09 & 10. Last night I loaded the 300 MHz design (but running at 200 MHz) for more tests by Jack. He thinks the design may run correctly (?) if we reduce the accumulation length to 2048 samples/accumulation. I will try that tomorrow. Meanwhile, I reloaded the 200 MHz design this morning, and will need to take calibration data to set delays. Observations on 2253+161 started a few minutes ago.
11:50 UT I reset the delays. They were very large again (mostly Ant 14), which is very strange to me. Interestingly, for ant14, the differences between delays several days ago and now are small, so there was a large change day before yesterday, but today's change is just as large, and in the opposite sense. However, other antennas had more complex changes, so this is not a general rule...
17:43 UT I loaded the 300 MHz correlator and switched over all of the software for fast-correlator operation. I set the accumulation length to 2048 (instead of the nominal 2560), and the x_acc_len to 3, which should allow the design to work without overrunning the slot boundaries. I took a bit of solar data, but then went to 3C84. Annoyingly, I see that I went there when it is in its unreachable window, so the data will only be good for ants 9-13 until another 20-30 minutes.
23:59 UT I switched back to the 200 MHz correlator about 18:40 UT, and we have been on normal solar schedule since then, but they were working on the 27-m preload cable, so the antenna is in LOCAL mode and cannot do calibrations. So I was not able to set the delays, and today's data will not be useful. I am not sure why the preload cable adjustment takes so long.
22:03 UT The 27-m is working again now. I finally had time to update the delays, and restarted a solar scan now. I will check the delays based on the calibrator scan coming up in a couple of hours.
19:00 UT We have decided to start taking all calibrator data in the all-band mode (pcal_hi-all.fsq), including both refcals at the start and end of each day, and the "daily" phase calibrations during the day. That necessitates somewhat longer scans of 1/2 hour each.
12:00 UT We have had a couple of weeks of nicely calibrated data (no flares to speak of, however). We are now working on the data pipeline, and have realized that secular variations in the calibrator phases require a bit more understanding. To that end, today we are taking data on 2136+006, 2253+161, and 0319+415 (3C84), using pcal_hi-all.fsq, alternating between sources for 20-min each.